BarneyBanano wrote:Well, Joe_H, thanks for the extremely rude and condescending reply! That's exactly what I was expecting from you after reading your replies to other posters. Thanks for not disappointing with your douchery! The posts I saw were from years ago, so I wasn't sure if there had been any progress. My apologies for wasting your precious time with a redundant question.
I noticed the documentation is pretty disorganized and has a lot of references to clicking links that are non-existent. For example -
https://foldingathome.org/support/faq/r ... ldinghome/ - under the question "Does Folding@home run on my graphics chip or GPU?", there is reference to "See the GPU and High Performance FAQs" but it is not linked anywhere, nor is there a heading that matches that in the navigation. Many more examples exist of this poor structure. In addition to the poor organization, why would the lack of GPU support on macOS not be mentioned under the above FAQ question? There are also multiple mentions of the use of GPU slots in the Mac Installation Guide -
https://foldingathome.org/support/faq/i ... uides/mac/. A little misleading, don't you think?
Donating folding power to your project is voluntary. I'm trying to help. I'm not an expert, I'm just a person who has a computer on 24/7 and would like it to be used for a good cause. Condescension isn't the best approach to encourage others to help with your project.
One factor you do not seem to grasp here, is that Joe_H is just another volunteer like you or I. It is not 'his' project any more than it is yours or mine. (He IS more invested in volunteering, and less on criticizing) The documentation is all by volunteers, you are free to improve it as you can. To improve the support here, just give that support yourself. I do.
There are topics that come up every week, (day, month) and the regulars get tired of giving the same answer each time. Burn out happens to us all.
A terse answer is a clear answer. My usual long-winded answers sometimes raise hope where no hope is to be found. A shorter answer may be more kind than to raise hope.
When F@H first decided to move to OpenCL from their earlier proprietary GPU interfaces, (CUDA, CAL, BROOK) the Apple implementation of OpenCL had a bug that prevented F@H processing. Once Apple fixed that bug, they started using Intel embedded GPUs. Very recently Apples with AMD (mostly) GPUs are being introduced again. Perhaps there will be a look in that direction again, but currently the number of potential donors is very small. Any investment in their limited resources should be aimed to produce the most scientific result. (Core_a7 does better, faster, Science than Core_a4, Core_22 does better Science than Core_21 ... except on OS-X) F@H appears to employ a single programmer, his time needs to be focused on projects with a important return. (Other development through the years has bee by sponsors, F@H does not pay for their programmers; Sony for the PS3 and Android clients, Nvidia for the CUDA client, AMD (ATI at the time) for the client using Brook)
Does any of that make you feel better than just saying "no"?
